From cbc799f4ea592dbb68c8ad859adec2888fdc5423 Mon Sep 17 00:00:00 2001
From: Grant Gainey <ggainey@redhat.com>
Date: Wed, 12 Jul 2023 14:29:48 -0400
Subject: [PATCH] Fixed content-handler response-headers/object-storage
 collision.

fixes #4028.
---
 CHANGES/4028.bugfix         |  1 +
 pulpcore/constants.py       | 37 +++++++++++++++++++++++++++++++++++++
 pulpcore/content/handler.py | 26 +++++++++++++++++---------
 3 files changed, 55 insertions(+), 9 deletions(-)
 create mode 100644 CHANGES/4028.bugfix

diff --git a/CHANGES/4028.bugfix b/CHANGES/4028.bugfix
new file mode 100644
index 00000000000..b9a2716fec7
--- /dev/null
+++ b/CHANGES/4028.bugfix
@@ -0,0 +1 @@
+Taught downloader to correctly handle plugin-specified headers for object-storage backends.
diff --git a/pulpcore/constants.py b/pulpcore/constants.py
index 48e490eeed8..80b7e46b6b4 100644
--- a/pulpcore/constants.py
+++ b/pulpcore/constants.py
@@ -67,3 +67,40 @@
 )
 
 EXPORT_BATCH_SIZE = 2000
+
+# Mapping of http-response-headers to what various block-storage-apis call them
+# https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/s3/client/get_object.html
+# response-headers S3 respects, and what they map to in an S3 object
+S3_RESPONSE_HEADER_MAP = {
+    "Content-Disposition": "ResponseContentDisposition",
+    "Content-Type": "ResponseContentType",
+    "Cache-Control": "ResponseCacheControl",
+    "Content-Language": "ResponseContentLanguage",
+    "Expires": "ResponseExpires",
+    "Content-Encoding": "ResponseContentEncoding",
+}
+# https://learn.microsoft.com/en-us/python/api/azure-storage-blob/azure.storage.blob.contentsettings?view=azure-python
+# response-headers azure respects, and what they map to in an azure object
+AZURE_RESPONSE_HEADER_MAP = {
+    "Content-Disposition": "content_disposition",
+    "Content-Type": "content_type",
+    "Cache-Control": "cache_control",
+    "Content-Language": "content_language",
+    "Content-Encoding": "content_encoding",
+}
+# https://gcloud.readthedocs.io/en/latest/storage-blobs.html
+# response-headers Google Cloud Storage respects, and what they map to in a GCS object
+GCS_RESPONSE_HEADER_MAP = {
+    "Content-Disposition": "content_disposition",
+    "Content-Type": "content_type",
+    "Cache-Control": "cache_control",
+    "Content-Language": "content_language",
+    "Content-Encoding": "content_encoding",
+}
+
+# Storage-type mapped to storage-response-map
+STORAGE_RESPONSE_MAP = {
+    "storages.backends.s3boto3.S3Boto3Storage": S3_RESPONSE_HEADER_MAP,
+    "storages.backends.azure_storage.AzureStorage": AZURE_RESPONSE_HEADER_MAP,
+    "storages.backends.gcloud.GoogleCloudStorage": GCS_RESPONSE_HEADER_MAP,
+}
diff --git a/pulpcore/content/handler.py b/pulpcore/content/handler.py
index 7a12301375e..3dd5743082e 100644
--- a/pulpcore/content/handler.py
+++ b/pulpcore/content/handler.py
@@ -20,6 +20,7 @@
 
 import django
 
+from pulpcore.constants import STORAGE_RESPONSE_MAP
 from pulpcore.responses import ArtifactResponse
 
 os.environ.setdefault("DJANGO_SETTINGS_MODULE", "pulpcore.app.settings")
@@ -894,6 +895,16 @@ async def _serve_content_artifact(self, content_artifact, headers, request):
         Returns:
             The :class:`aiohttp.web.FileResponse` for the file.
         """
+
+        def _set_params_from_headers(hdrs, storage_domain):
+            # Map standard-response-headers to storage-object-specific keys
+            params = {}
+            if storage_domain in STORAGE_RESPONSE_MAP:
+                for a_key in STORAGE_RESPONSE_MAP[storage_domain]:
+                    if hdrs.get(a_key, None):
+                        params[STORAGE_RESPONSE_MAP[storage_domain][a_key]] = hdrs[a_key]
+            return params
+
         artifact_file = content_artifact.artifact.file
         artifact_name = artifact_file.name
         filename = os.path.basename(content_artifact.relative_path)
@@ -909,9 +920,8 @@ async def _serve_content_artifact(self, content_artifact, headers, request):
         elif not domain.redirect_to_object_storage:
             return ArtifactResponse(content_artifact.artifact, headers=headers)
         elif domain.storage_class == "storages.backends.s3boto3.S3Boto3Storage":
-            parameters = {"ResponseContentDisposition": content_disposition}
-            if headers.get("Content-Type"):
-                parameters["ResponseContentType"] = headers.get("Content-Type")
+            headers["Content-Disposition"] = content_disposition
+            parameters = _set_params_from_headers(headers, domain.storage_class)
             url = URL(
                 artifact_file.storage.url(
                     artifact_name, parameters=parameters, http_method=request.method
@@ -920,15 +930,13 @@ async def _serve_content_artifact(self, content_artifact, headers, request):
             )
             raise HTTPFound(url)
         elif domain.storage_class == "storages.backends.azure_storage.AzureStorage":
-            parameters = {"content_disposition": content_disposition}
-            if headers.get("Content-Type"):
-                parameters["content_type"] = headers.get("Content-Type")
+            headers["Content-Disposition"] = content_disposition
+            parameters = _set_params_from_headers(headers, domain.storage_class)
             url = URL(artifact_file.storage.url(artifact_name, parameters=parameters), encoded=True)
             raise HTTPFound(url)
         elif domain.storage_class == "storages.backends.gcloud.GoogleCloudStorage":
-            parameters = {"response_disposition": content_disposition}
-            if headers.get("Content-Type"):
-                parameters["content_type"] = headers.get("Content-Type")
+            headers["Content-Disposition"] = content_disposition
+            parameters = _set_params_from_headers(headers, domain.storage_class)
             url = URL(artifact_file.storage.url(artifact_name, parameters=parameters), encoded=True)
             raise HTTPFound(url)
         else: